CN103329527A - Compression and decompression of reference images in a video coding device - Google Patents
Compression and decompression of reference images in a video coding device Download PDFInfo
- Publication number
- CN103329527A CN103329527A CN2012800052715A CN201280005271A CN103329527A CN 103329527 A CN103329527 A CN 103329527A CN 2012800052715 A CN2012800052715 A CN 2012800052715A CN 201280005271 A CN201280005271 A CN 201280005271A CN 103329527 A CN103329527 A CN 103329527A
- Authority
- CN
- China
- Prior art keywords
- image
- regulation
- coding mode
- reconstructed
- intra
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/42—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation
- H04N19/423—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation characterised by memory arrangements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/40—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using video transcoding, i.e. partial or full decoding of a coded input stream followed by re-encoding of the decoded output stream
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
- H04N19/146—Data rate or code amount at the encoder output
- H04N19/147—Data rate or code amount at the encoder output according to rate distortion criteria
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/117—Filters, e.g. for pre-processing or post-processing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/124—Quantisation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
- H04N19/157—Assigned coding mode, i.e. the coding mode being predefined or preselected to be further used for selection of another element or parameter
- H04N19/159—Prediction type, e.g. intra-frame, inter-frame or bidirectional frame prediction
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/42—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation
- H04N19/423—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation characterised by memory arrangements
- H04N19/426—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation characterised by memory arrangements using memory downsizing methods
- H04N19/428—Recompression, e.g. by spatial or temporal decimation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/46—Embedding additional information in the video signal during the compression process
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
本发明涉及用于在预测编码的范围中提供重构图像的方法和设备。在此情况下重构图像按照图像编码设备的编码模式以压缩的方式被存放在参考图像存储器中。
The invention relates to a method and a device for providing reconstructed images in the context of predictive coding. In this case the reconstructed image is stored compressed in the reference image memory according to the encoding mode of the image encoding device.
Description
Technical field
The present invention relates to be used to the method and apparatus that reconstructed image is provided.
Background technology
The picture format that will encode in the past few years is for example owing to the introducing of Novel shooting system becomes increasing.Thereby especially occur at present from full images size with 625x576 picture point nearest 50 years in Europe employed television transmission system PAL(line-by-line inversion method (Phase Alternation Line)) to the HDTV(high definition TV with 1920x1080 picture point or 1280x720 picture point) and conversion.What expect is also will introduce larger picture format in the Novel television system in the future.
HDTV and system in the future use digital compression techniques, so as with the sequence of video images boil down to so that these video images for example can transmit via the Internet or via mobile radio channel.But because the amplification of picture format significantly increases for the needed rated output of compressed video image sequence with at this needed storage requirement.Its consequence is that the data transmission between the computing unit of memory and realization compression method also significantly raises.
Therefore such as Video coding integration and cooperation group (Joint Colloborative Team on Video Coding, JCT-VC), the co-operation group of ITU and ISO/IEC (ITU-International Telecommunication Union(International Telecommunications Union), ISO-International Standardization Organisation(International Standards Organization), IEC-International Electrotechnical Commission(International Electrotechnical Commissio)) working group not only is engaged in the improvement of compression ratio, and be engaged in standardized method, in order to can efficiently video image be left in the reference picture store of corresponding codec and can conduct interviews in the mode of saving resource.
Fig. 1 illustrates for the known device with reference picture store SRB that image sequence is compressed.In the case, image is for example by means of also known predictive coding is encoded as the INTER coding mode.One of image is broken down into the image block BB that for example has 16x16 picture point and then is encoded block by block.Then seek a reference image block RBB for one of image block in reference image R EF, described reference image block guarantees the good basis for the picture material of estimating this image block.This image block is submitted to motion estimation unit ME for this reason, described motion estimation unit is selected this reference image block according to reference subimage REFT from the reference subimage, and selected reference image block is signaled the wherein said part that comprises the reference image R EF after carrying out image decompressor by image decompressor unit PC with reference to subimage to motion compensation units MC by means of motion vector MV.This motion compensation units provides described reference image block based on this reference picture and this motion vector.
In next step, produce difference image piece BD by from image block BB, deducting reference image block RBB.Make subsequently the difference image piece in converter unit T, for example stand conversion according to discrete cosine transform method.Output at this converter unit provides conversion coefficient TK, and described conversion coefficient is fed to quantifying unit Q subsequently in order to quantize.Provide conversion coefficient TQ through quantizing at the output of this quantifying unit, these conversion coefficients through quantizing are by being transformed into output signal AS by the performed entropy coding of entropy coding unit EC.
In feedback loop, the conversion coefficient TQ through quantizing is transformed into the conversion coefficient TKR of reconstruct by means of the re-quantization by inverse quantization unit IQ.The conversion coefficient TKR of these reconstruct is transformed into the difference image piece BDR of reconstruct by the inverse transformation of inverse transformation block IT.In another step, by the difference image piece BDR of reconstruct and reference image block RBB addition being generated the image block RBM of reconstruct.
In older coding method, the image block of reconstruct directly is written in the reference picture store.Be in the standardized method current, make the image block of reconstruct at first also stand image compression by image compression unit PC in order to reduce data capacity, described image compression obviously reduces the data capacity of the image block of reconstruct.The compressed reconstructed image piece RBC that forms by image compression unit PC is stored in the reference picture store subsequently.In order to allow motion estimation unit and motion compensation units to access needed view data, when the specific image fragment of request reference image R EF or this reference picture, at first from reference picture store SRB, read corresponding compressed reconstructed image piece and be transformed into reference to subimage REFT by means of the image decompressor by image decompressor unit PD.
Fig. 2 illustrate with at the corresponding decoder of the encoder shown in Fig. 1.In the case, output signal AS is decoded into conversion coefficient TQ through quantizing by means of entropy decoding unit ED.In addition, the conversion coefficient through quantizing is become the conversion coefficient TKR of reconstruct by re-quantization by means of inverse transformation block IQ.Then by contrary information unit IT the conversion coefficient TKR inversion of reconstruct is changed into the difference image piece BDR of reconstruct.
Except this output signal, especially also send corresponding motion vector MV to decoder.Decoder is in the situation that use and can determine reference image block RBB by motion compensation units MC thus with reference to subimage REFT, and described reference image block is by being transformed into the image block RBM of reconstruct with the difference image piece addition of reconstruct.
The image block RBM of reconstruct for example can be forwarded to display.The image block RBM of reconstruct is transformed into compressed reconstructed image piece RBC by means of the compression by image compression unit PC subsequently, and this compressed reconstructed image piece then is stored among the reference picture store SRB.Leaving compressed reconstructed image piece in this reference picture store in can be extracted by image decompressor unit PD and shorten into reference to subimage.
Document [1] has been described Lossless Image Compression method/image decompression compression method, wherein in floating-point dct transform (DCT-Discrete Cosinus Transformation(discrete cosine transform)) and scanned the coefficient of two-dimensional arrangement after the conversion after encode with one-dimensional representation execute bit aspect.
A kind of memory access bandwidth of suggestion reduces technology in according to the method for document [2].In the case except also advising that for image compression unit PC DC predicts and entropy is encoded or advise the therewith step of contrary for image decompressor unit PD after the transform and quantization.
In another document [3], introduce the test result before the image storage of deblocking with afterwards in compression or decompressed image data the time.
In document [4], introduce at last the image storage compacting method (DPCM-Diskrete Pulscode Modulation (discrete pulse coded modulation)) based on one dimension DPCM.
The compression method of advising in document [1] and [4] at least is break-even.
Summary of the invention
The task of having explanation following methods and equipment utilizes described method and apparatus to realize implementing for the resource high-efficiency that the reconstructed image piece is carried out compression/de-compression for image encoding apparatus and/or image decoding apparatus.
This task solves by independent claims.Improvement project of the present invention can draw from dependent claims.
The present invention relates to a kind of method, at least one image area that is used for the storage reconstructed image, described reconstructed image generates by image encoding apparatus or by image decoding apparatus, with be used to image encoding apparatus or image decoding apparatus to provide at least one image area as the reference image, have following steps:
A) based on the first regulation at least one image area is compressed into compressed data;
B) the compressed data of storage;
C) based on the second regulation (V2) compressed data (KDAT) are de-compressed at least one image area (BR);
It is characterized in that,
D) form the first regulation (V1) based on the 3rd regulation (V3), wherein the 3rd regulation (V3) defines the coding mode (M1) of image encoding apparatus (ENC),
E) form in the following manner the second regulation (V2), namely the second regulation (V2) is selected as the 3rd contrary regulation (V3I) with the 3rd regulation (V3) contrary.
The present invention can realize the depositing of memory-efficient of reference picture, and described reference picture is required in the coding by image encoding apparatus or the decoding scope by image decoding apparatus.Advantageously, based on one of coding mode of image encoding apparatus to reconstructed image, also be that reference picture compresses.Especially can realize thus being used for compressing or the recessiveness regulation of the first or second regulation of decompressing for establishment.In addition, often follow rate-distortion optimisation for the coding decision of the image encoding apparatus that image block is encoded, this rate-distortion optimisation expends optimized image quality in the situation at minimized memory.So this best coding mode is formed for the basis that the image area to reconstructed image compresses, and also can obtain high compression rate thus.In addition, a lot of at the hardware servicing unit of equipment support available on the market for the treatment of the substep that in image encoding apparatus or image decoding apparatus, is performed at least.Thereby the compression of reference picture or decompression also can be used this hardware servicing unit.
In present-day standard image encoding apparatus and image decoding apparatus, before the storage reconstructed image of being everlasting, make described reconstructed image stand de-blocking filter and/or other filtering operation, for example Wiener filtering.In a kind of improvement project of the present invention, after decompressing in the use situation of the present invention, just carry out these steps.Can use by compression thus can be for coding mode, the compression ratio of coefficient in order to be improved conversion or quantification in the respective image district of reconstructed image.
Preferably, compressed data are stored in the reference picture store with index, and wherein this index comprises for the sign that identifies clearly the first regulation or the second regulation.Also can use thus for compression or the pattern that decompresses, described pattern can be determined and with respect to can the recessive pattern of determining obtaining better compression ratio by non-recessively.
In a kind of improvement project of the present invention, according to storing encoded data for generation of the grammer of output stream, described output stream (i) produces and/or (ii) receives by image decoding apparatus by image encoding apparatus.Thus, image decoding apparatus acquired information: image encoding apparatus has as far as possible nondestructively carried out compression and decompression to each image area of reconstructed image in which way.Can avoid thus the drift between image encoding apparatus and image decoding apparatus.
In favourable improvement project, a plurality of image areas of reconstructed image are unified into an image-region and are stored as and can irrespectively decompress with other image-region of same reconstructed image.Realized thus: this image-region of reconstructed image can be irrespectively accessed with other image-region of same reconstructed image.The calculating that reduce to be used for decompresses thus expends and to access time of the picture material of image-region, because needn't the picture material of other image-region be decompressed except the image-region of asking.
Advantageously, the corresponding image area of image-region is according to the first unique regulation or compressed according to corresponding the first regulation sequence.Expend thereby reduce thus expending with the minimizing memory of notice, because needn't in reference picture store, deposit the first or second independent regulation for each image area.
Advantageously, at least some image areas of image-region come compressed by the INTRA predictive mode, wherein also will store except the compressed data of storing this image-region at least one picture point to this image-region and be reconstructed the first picture point needed and that can only determine by the image-region that is different from this image-region.Usefulness to the INTRA predictive mode can be carried out simply.
If the first picture point is according to regulation that can be given in advance, especially compressed according to the Delta coding and be stored, the memory location that then is used for the memory image district is further reduced in the mode by means of the compression of INTRA predictive mode.
In favourable improvement project of the present invention, at least one image area of image-region be not by the INTRA predictive mode but by the INTRA coding mode compressed.Here for example also can compress efficiently because picture noise or strenuous exercise and the image area that can be predicted poorly.
Advantageously, from a plurality of (i) by selecting following pattern at least two different INTRA predictive modes or the pattern that (ii) consisted of by at least one INTRA predictive mode and INTRA coding mode: this pattern has minimum memory location demand for the image area that will compress.Realized by this optimization: reconstructed image is can be with good quality decompressed and simultaneously can be compressed with high compression ratio.
If preferably storage is used for the quantization parameter that INTRA predicts under the INTRA prediction case, then can guarantee high-quality decompression.
In a kind of improvement project of the present invention, store at least two reconstructed images, wherein the first reconstructed image at least two reconstructed images is by by means of INTRA predictive mode and/or the compression of INTRA coding mode, and second reconstructed image of following in time the first reconstructed image at least two reconstructed images is by by means of the compression of INTER coding mode, and the motion vector of described INTER coding mode is with reference to the first reconstructed image.Can realize with respect to pure INTRA prediction thus the raising of compression ratio.
Preferably, produce the first regulation based on the 3rd regulation by one of following steps:
If the-the three regulation is described the INTRA coding mode, then the first regulation consists of by identical INTRA coding mode;
If the-the three regulation is described following coding mode, macro block/image block that (i) will encode in this coding mode then selects to be used for INTRA coding mode that reference picture is encoded for the first regulation by " jump (skip) " signal, also namely in the situation that the reference picture of the macro block/image block that is encoded and (ii) will encodes without motion vector is encoded by means of the INTRA coding mode;
If the-the three regulation is used the INTER coding mode with affiliated motion vector for image block/macro block, then stipulate to select in the same manner the first regulation with the 3rd, wherein select in the same manner to be used for the motion vector of the 3rd regulation and the first INTER coding mode of stipulating.
Can realize notifying for the recessive signal that first or second of compression or decompression stipulated by using one of these regulations.In addition, can again the coding mode of having optimized be used for compression, can avoid thus expending for the calculating of determining best the first regulation.
In the favourable improvement project of the method according to this invention, according to one of aforementioned enforcement, wherein for synchronous the method in image encoding apparatus and in image decoding apparatus, notifying one of at least following parameter between image encoding apparatus and image decoding apparatus:
-for the index of the image-region of the image area of reconstructed image or reconstructed image, wherein said index comprises for the sign that identifies clearly the first regulation or the second regulation;
-quantizing factor, it is used at least one image area is compressed in INTRA predictive mode or INTRA coding mode situation.
Can improve compression ratio thus, because needn't only carry out compression according to the first regulation that can recessively notify.Therefore, exist the more big collection of the 3rd regulation available, these the 3rd regulations can be considered by the first regulation.
Except described regulation, can determine the first and second regulations according in the following behavior at least one:
A) use the different compression method of implementing by image compression unit or image decompressor unit:
-in output signal to the linearity of employed method notice;
-compression method distributed to respectively be used to configuration file (Profilen) and profile/level that the image block of image encoding apparatus or image decoding apparatus is encoded;
B) parameter of description compression method:
-parameter set is to the distribution of profile/level combination;
-parameter is in band or as the notice of independent parameter set/transmission;
-reduce based on for example parameter at the data available of the quantization parameter of front image, current use;
C) for the possibility that reduces parameter based on data available:
-the statistics carried secretly
-for the image section of being encoded by INTRA:
-) adopt current coding mode, perhaps
-) pattern is changed over accordingly or very similar INTRA compact model;
-) derivation prediction direction (being used by other compact model)
-by especially quantizing other coding step derivation parameter of (quantization step or quantization parameter), map function and filtering operation.
This derivation can comprise the digital space conversion, make the value of calculating ad hoc become integer, also comprises the formula for direct calculating parameter.
If parameter is derived not based on the data that can use at decoder-side, be possible below then:
The statistics of-definite image that will compress;
-use other supplementary, for example sensor information;
-multipath coding (Multipath-Encodierung)--come Optimal Parameters (also being suitable for the selection of compression method) by considering a plurality of images or complete sequence;
The section of-identification scene, for example image needn't be compressed, because described image is not used as reference picture;
-use special available parameter to be used for long term reference image (=long-time reference picture), for example very high quality can can't harm;
Condition:
The data rate restriction of-transmission channel allows to use more coarse parameter;
Quality limit on the-image aspect requires based on statistical scheme, carries out in the mode of block-by-block because process;
-the quality limit on the piece aspect can be directly involved in the quality of the piece that will compress;
Parameter is derived and the potential limit of process choosing as being used for for the data of-the minimum mass that realize, the maximum quantity of generation and inner data/address bus throughput;
Compression is used for reference information causes different memory script (=memory mode or memory requirement) for each reference picture.This is particularly by using different compression processes and drawing with dynamical fashion for the parameter of compression process by deriving.
-memory organization (for example view data being divided into subarea=definite size) thus determine that memory fully uses and determine the quantity of reference picture;
-with reference to parameter configuration file/level combination is distributed in storage management, subarea size for example.
The advantage of these behaviors: need the notice of parameter and/or derive to be used for for subsequent process step, memory-size and internal data bus throughput and the reference information of encoder of the identical behavior of compression process require optimally to be controlled at to(for) picture quality.
The invention still further relates to a kind of equipment, at least one image area that is used for the storage reconstructed image, described reconstructed image generates by image encoding apparatus or by image decoding apparatus, and be used to image encoding apparatus or image decoding apparatus to provide at least one image area as the reference image, have with lower unit:
A) for the image compression unit that at least one image area is compressed into compressed data based on the first regulation;
B) be used for depositing together compressed data storage unit;
C) for the image decompressor unit that compressed data decompression is shortened at least one image area based on the second regulation;
It is characterized in that,
D) form the first regulation based on the 3rd regulation, the coding mode of the 3rd stipulative definition image encoding apparatus wherein,
E) form in the following manner the second regulation, namely the second regulation is the 3rd contrary regulation with the 3rd regulation contrary.
This equipment has processing unit in addition, and described processing unit is constituted as at least one step that makes it possible to achieve and implement the method.
The advantage of this equipment is similar to the advantage of described method.
At last, be a part of the present invention according to the method for above-mentioned steps or according to the application of equipment in image encoding apparatus (ENC) and/or image decoding apparatus of above-mentioned enforcement.Can realize in an advantageous manner and implement thus the present invention and its improvement project.
Description of drawings
With reference to the accompanying drawings the present invention and its improvement project are set forth in more detail.
Particularly:
Fig. 1 illustrate by prior art known be used for encoder that view data is compressed;
Fig. 2 illustrate by prior art known be used for decoder that compressed view data is decompressed;
Fig. 3 illustrates the first embodiment of image encoding apparatus;
Fig. 4 illustrates the first embodiment of image decoding apparatus;
Fig. 5 a and 5b illustrate according to standard two INTRA predictive modes H.264/AVC;
Fig. 6 illustrates for compressed and picture point uncompressed that leaves reference picture store in;
Fig. 7 illustrates the other embodiment with two reference pictures;
In the drawings, the element that has identical function and a type of action is equipped with identical Reference numeral.
Embodiment
Fig. 1 and 2 illustrates the encoder that is used for encoding/decoding image according to prior art.Fig. 1 and 2 is at large set forth in foreword, thereby herein with reference to the narration in foreword.
The below for example is interpreted as reconstructed image piece RBC with at least one image area BR of reconstructed image RBD.In addition, at least one image area with reference to image for example is interpreted as reference image block.In addition, reference picture store is also referred to as memory cell SRB.
Usually, encoder is used reference picture store, to realize different predictive coding patterns.The coding mode of introducing in Fig. 1 and 2 is the INTER coding mode, wherein before has been encoded in time and the reconstructed image content of the image of decoding or reconstruct is used to predict the current image that will encode and maybe will decode.In the prior art, a lot of predictive coding patterns or corresponding decoding schema have been described, so that this is not further inquired into.
Be also referred to as the image encoding apparatus ENC of encoder below Fig. 3 is schematically illustrated, described encoder produces output signal AS from the image with image block BB.This functional mode is similar to the functional mode according to the equipment of Fig. 1.Here use the synonym image area BR of reconstructed image RBD as reconstructed image piece RBM, wherein reconstructed image comprises that all belong to the reconstructed image piece of this image, and uses the compressed data KDAT of synonym as compressed reconstructed image piece RBC.
Selected cell ASE selects following coding mode from a plurality of coding mode M1, M2, this coding mode in the situation that picture quality can given in advancely minimize can be given in advance for the memory location demand of the memory location demand of storage output signal or the respective image piece after being used for memory encoding situation under maximize the picture quality of the reconstructed image piece corresponding with image block.Among coding mode M1, the M2 each represents respectively one of the 3rd possible regulation V3, the coding of its Description Image piece.Similarly, there are respectively the 3rd contrary regulation V3I with the 3rd regulation V3 contrary, the decoding of the data block that its definition is encoded among coding mode M1, the M2 each.
Stipulate the example of V3 according to the 3rd of Fig. 1:
-selection reference image block RBB from reference subimage REFT;
-produce difference image piece BD by from the image block BB of image, deducting this reference image block;
-from difference image piece BD, produce conversion coefficient TK by cosine transform;
-by quantizing from conversion coefficient TH, to produce the conversion coefficient TQ through quantizing;
Example according to the 3rd contrary regulation V3I Fig. 1 or 2 and the 3rd regulation contrary:
-by the conversion coefficient TKR of re-quantization from generation reconstruct through the conversion coefficient TQ of quantification;
-from the conversion coefficient TKR of reconstruct, produce the difference image piece BDR of reconstruct by inverse transformation;
-by produce the image block RBM of reconstruct with reference to the difference image piece BDR addition of image block RBB and reconstruct.
As shown in FIG. 3, the three regulation V3 related with selected coding mode is submitted to image compression unit PC as the first regulation.This image compression unit is compressed according to the image block RBM of this first regulation to reconstruct, described the first regulation or identical with the 3rd regulation, or by the 3rd regulation derivation.Example to this is:
If the-the three regulation is described the INTRA coding mode, then can use identical INTRA coding mode for the first regulation;
If the-the three regulation is described following coding mode, macro block/image block that (i) will encode in this coding mode by " jump (skip) " signal and in the situation that without motion vector be encoded and (ii) reference picture be encoded by means of the INTRA coding mode, then provide following coding mode as the first regulation, in this coding mode, use the macro block/image block that to encode by means of the INTRA coding mode that is used for the reference picture coding;
If the-the three regulation is used the INTER coding mode with affiliated motion vector for image block/macro block, then for compressing, the image block/macro block to reconstruct adopt the 3rd regulation as the first regulation, also namely compress for the image block of counterweight structure, image compression unit is also used the INTER coding mode and identical motion vector according to the 3rd regulation.
Compress the compressed data KDAT of generation by the image block RBM to reconstruct.These compressed data utilize position description APOS to deposit, and the image block that described position description reproduces reconstruct is in reference picture or the position in the reconstructed image RBD of the image block of describing reconstruct.Can find encoded unit thus, such as the image block of reconstruct.In order to access the image area of determining in the reference picture, form TAB is provided in reference picture store, described form shows, can find at which memory location the image block that belongs to reconstruct in or to belong to the compressed data of the image area in the reconstructed image.If compressed data can not realize the clear and definite decompression by decompression unit, enclose index INX and leave in the reference picture store then can for described compressed data, wherein index INX comprises for the sign that identifies clearly the first regulation and/or the second regulation.Therefore, decompression unit can be determined the second regulation V2 and determine reference picture part REFT from compressed data KDAT according to index INX.Replacedly, this index can be stored among the form TAB.
Image compression unit PC, image decompressor unit PD and reference picture store SRB can united be the compression module XC that is also referred to as equipment SV.
In the above example, can from the 3rd regulation, directly obtain the first regulation.Be not in the unique situation in the first regulation to the distribution of the 3rd regulation, can additionally insert the information field with index INX in output signal, described information field is notified the first regulation dominantly.
Fig. 4 illustrates for the image decompressor cells D EC that is called below decoder and uses behavior of the present invention.Output signal is received and processes with being similar to Fig. 2.Be similar to Fig. 3 ground and use concept.Compression module with its unit: image compression unit PC, image decompressor unit PD and reference picture store SRB is similar to the description work among Fig. 3.Different from the encoder according to Fig. 3, in the decoder situation, dependently coding mode is selected.On the contrary, perhaps by means of the information field in the output signal the 3rd regulation is notified to decoder dominantly, for example by means of index INX, perhaps dominant the special configuration of the 3rd regulation given in advance in encoder, perhaps decoder can be from output signal for example be derived the 3rd regulation according to the coding mode or the picture material that are used for image block is encoded far and away.So can from the 3rd regulation, derive clearly the 3rd contrary regulation.When encoder comes that with identical the 3rd regulation or the 3rd contrary regulation identical image block carried out compression and decompression, be very useful.Otherwise the pseudo-shadow of encoding may appear.
In another embodiment, use the INTRA prediction by image compression unit or image decompressor unit.The INTRA prediction is characterised in that, predict the image block that will compress and the error signal that draws together with the information about prediction by means of the image segments of the reconstruct in the image identical with the image block that will compress, for example predictive mode or motion vector are encoded together.H.264/AVC, standard defines nine different INTRA predictive modes, and it the following is:
The MODE name (with English and Chinese) of INTRA prediction
Pattern 0:Vertical(is vertical)
Pattern 1: level
Pattern 2:DC pattern (same components pattern)
Mode 3: Diagonal-Down-Left(diagonal lower-left)
Pattern 4:Diagonal-Down-Right(diagonal bottom right)
Pattern 5:Vertical-Right(vertical right)
Under the pattern 6:Horizontal-Down(level)
Mode 7: the Vertical-Left(vertical left)
On the pattern 8:Horizontal-Up(level).
Exemplarily in Fig. 5 a, describe pattern 0 and in Fig. 5 b, described pattern 1.At this, in the 4x4 of runic edge image block, can find out the reconstructed image piece with 4x4 picture point.For application model 0, need to be positioned at 4 picture points of image block of reconstruct of the image block top of current reconstruct, referring to Reference numeral X.Similarly, four picture points of image block of reconstruct that need to be positioned at the image block left side of current reconstruct in pattern 1 are used for compression/de-compression.
For the image to reconstruct compresses, the image of this reconstruct is divided into image-region, and wherein these image-regions are irrespectively compressed each other.Can access so thus the picture material of reference picture, so that only need the picture material of one of image-region is processed.In the behavior situation that is used for compression, there are a plurality of flexible programs.
Flexible program 1:
In the first flexible program, the picture point that can not be reconstructed in corresponding image-region is non-to be stored in the reference picture store compressively.Therefore for example will be located immediately under only using pattern 0 situation that whole picture point row above the reconstructed image piece that will compress is non-to be left in the reference picture store compressively.Fig. 6 shows the image-region with 3x5 reconstructed image piece.In the case, the unpressed picture point of row representative that horizontal stripe is arranged above this image-region.
The picture point that usually, can generate by other zone of reconstructed image in flexible program 1 situation does not leave in the reference picture store compressively.Can access thus the image area of determining in the image-region that is present in the mode of compressing in the reference picture store.
Flexible program 2
In the variant of flexible program 1, unpressed picture point is stored in the mode of compression at least in part.For this reason, all picture points can be according to can regulation given in advance being set up to a row, and compress by means of Delta coding or other poor coding.
Example:
Picture point value: 100,80,115
Value after poor coding: 100 ,-20,15
In the case, the first picture point value is not encoded, and picture point value difference subsequently transmits with dividing.
Flexible program 3:
Can there be a plurality of INTRA predictive modes, for example in situation H.264/AVC.So can select specific pattern for each image-region, described specific pattern is applied to all reconstructed image pieces of this image-region.Replacedly, also can determine can be given in advance INTRA predictive mode sequence reconstructed image piece sequence is compressed being used for, for example:
Pattern 1-〉pattern 0-〉pattern 1-〉pattern 0-〉....
In the case, the first reconstructed image piece is compressed by means of pattern 1, and the second reconstructed image piece is compressed by means of pattern 0, and the reconstructed image block is compressed etc. by means of pattern 1.The sequence of these patterns can be stored in the reference picture store, in order to realize the personalization of compression.
Flexible program 4:
Except the INTRA predictive mode, also can there be the INTRA coding mode, it is encoded to the image block without reference information.Thereby the INTRA coding mode is called as pattern 9 in this flexible program, except INTRA predictive mode 0-8.
Determine in first step, in the pattern 0 to 9 which have the minimum memory locational requirement in the storage requirement situation of considering to be used for the informed code pattern and being used for the needed picture point of pattern that can not determine by image-region, corresponding is stored.
In second step, by based on the pattern with minimum memory locational requirement corresponding reconstructed image piece being compressed compressed data and is stored in reference picture store with the needed information that is used for decompressing.Explanation about pattern and picture point that can not be by image-region reconstruct belongs to this information.
Because in this flexible program, also consider the INTRA coding mode, be minimized so be used for storing compressed data storage locational requirement.
Flexible program 5:
In another variant, image compression unit also can be carried out the data that produce by INTRA prediction and/or INTRA coding mode and quantize.If quantizing factor can not be determined recessively, then this quantizing factor is stored in the reference picture store.
Flexible program 6:
Another variant can be, image compression unit for example with the form of section (Slice) produce be used for image-region with data flow operating such and will leave reference picture store in the data flow of operating such.
Above-mentioned flexible program can be combined at least in part.For fear of or be minimized in drift between the encoder, the information in the reference picture store of also being stored in except compressed data in corresponding flexible program can be inserted in the output signal and from encoder and be transferred to decoder.
In another embodiment, in reference picture store, deposit respectively two reference pictures.The first reference picture is stored in the reference picture store by means of the mode that INTRA predicts to compress.The second reference picture is stored in the reference picture store by means of the INTER coding mode in the mode of compression.In the case, seek reference image block in first reference picture of image block in leaving reference picture store in for each reconstruct.Preferably, when this reference image block is sought, motion vector is chosen as zero, so that except compressed data, also have only a small amount of information to be stored in the reference picture store.Image compression unit in the encoder or image decompressor unit also must compress corresponding reference picture or decompress here in an identical manner.In the distortion of this flexible program, the even number reconstructed image is stored in the reference picture store by means of the mode that INTRA predicts to compress, and the odd number reconstructed image is stored in the reference picture store by means of the mode of INTER coding mode with compression.
In another embodiment, referring to Fig. 7, can from two reference pictures (all prediction is stored in reference picture store as INTRA for both), select respectively one with reference to subimage REFT1, REFT2.In this diagram, dashed region C illustrates the incompressible reconstructed image data in the memory.Therefore at least some picture point X of reconstructed image piece for reconstructed image piece to be processed is subsequently compressed by intermediate storage in memory.In order to access simply the subarea with reference to subimage, for example for motion estimation unit ME or motion compensation units MC, in memory, provide at least a part as with reference to subimage REFT1, REFT2 in the reference picture each by corresponding image decompressor unit PD1, PD2.In Fig. 7, in memory, deposit macro-block line ML0, MB1 and MB2 for each with reference to subimage.Except the possibility that unique example is provided for the image decompressor unit, advantageously also can prepare Multi-instance, for example PD1, PD2.This has advantage: can be decompressed concurrently with reference to subimage accordingly.If no longer need the view data determined by corresponding image decompressor unit or image compression unit, then can from memory, delete these view data.This deletion can carried out as sliding window (Sliding Window) in a kind of configuration.
A plurality of reference pictures that should leave in the reference picture store can be notified between encoder.
Bibliography:
[1] JCTVC-B103: " Reference frame compression using image coder ", ISO/IEC Document:JCTVC-Bl03,2nd Meeting:Geneva, CH, 21-27 July, 2010, author: Chong Soon Lim
[2] JCTVC-B089: " Compressed Reference Frame Buffers (CRFB) ", ISO/IEC Document:JCTVC-B089,2nd Meeting:Geneva, CH, 21-28 July, 2010, author: the people such as Mehmet Umut Demircin
[3] JCTVC-B090: " ALF memory compression and IBDI/ALF coding efficiency test results in TMuC-0.1 ", JSO/IEC, Document:JCTVC-B090, author: Madhukar Budagavi
[4] JCTVC-B057: " DPCM-based memory compression ", ISO/IEC Document JCTVC-B057,2nd Meeting:Geneva, CH, 21-28 July 2010. authors: Hirofumi Aoki.
Claims (16)
1. at least one image area (BR) that is used for storage reconstructed image (RBD), described reconstructed image generates by image encoding apparatus (ENC) or by image decoding apparatus (DEC), with be used to image encoding apparatus (ENC) or image decoding apparatus (DEC) to provide at least one image area (BR) as with reference to the method for image (REF), have following steps:
A) based on the first regulation (V1) at least one image area (BR) is compressed into compressed data (KDAT);
B) the compressed data (KDAT) of storage;
C) based on the second regulation (V2) compressed data (KDAT) are de-compressed at least one image area (BR);
It is characterized in that,
D) form the first regulation (V1) based on the 3rd regulation (V3), wherein the 3rd regulation (V3) defines the coding mode (M1) of image encoding apparatus (ENC),
E) form in the following manner the second regulation (V2), namely the second regulation (V2) is selected as the 3rd contrary regulation (V3I) with the 3rd regulation (V3) contrary.
2. method according to claim 1, wherein compressed data (KDAT) are stored in the reference picture store with index (INX), and wherein this index (INX) comprises for the sign that identifies clearly the first regulation (V1) or the second regulation (V2).
3. method according to claim 1, wherein according to store encoded data (KDAT) for generation of the grammer of output stream, described output stream (i) produces and/or (ii) receives by image decoding apparatus (DEC) by image encoding apparatus (ENC).
4. according to the described method of one of aforementioned claim, wherein a plurality of image areas (BR) with reconstructed image (RBD) are unified into an image-region and are stored as and can irrespectively decompress with other image-region of same reconstructed image (RBD).
5. method according to claim 4, wherein the corresponding image area (BR) of this image-region is according to the first unique regulation (V1) or compressed according to corresponding first regulation (V1) sequence.
6. according to claim 4 or 5 described methods, wherein at least some image areas (BR) of this image-region come compressedly by the INTRA predictive mode, and wherein also storage (i) is needed and (ii) can only be by the first definite picture point of image-region that is different from this image-region at least one picture point to image-region is reconstructed except the compressed data (KDAT) of storing this image-region.
7. method according to claim 6, wherein the first picture point according to the rules, especially compressed according to the Delta coding and be stored.
8. according to claim 6 or one of 7 described methods, wherein at least one image area (BR) of this image-region is not by the INTRA predictive mode but compressed by the INTRA coding mode.
9. according to claim 6 or one of 7 described methods, wherein from a plurality of (i) by selecting following pattern at least two different INTRA predictive modes or the pattern that (ii) consisted of by at least one INTRA predictive mode and INTRA coding mode: this pattern has minimum memory location demand for the image area that will compress (BR).
10. according to claim 4 to one of 9 described methods, wherein storage is used for the quantization parameter of INTRA prediction.
11. according to claim 1 to one of 3 described methods, storage at least two reconstructed images (RBD) wherein,
The first reconstructed image at least two reconstructed images is by by means of INTRA predictive mode and/or the compression of INTRA coding mode, and second reconstructed image of following in time the first reconstructed image at least two reconstructed images is by by means of the compression of INTER coding mode, and the motion vector of described INTER coding mode is with reference to the first reconstructed image.
12. according to claim 1 to one of 3 described methods, wherein produce the first regulation (V1) based on the 3rd regulation (V3) by one of following steps:
If the-the three regulation is described the INTRA coding mode, then the first regulation consists of by identical INTRA coding mode;
If the-the three regulation is described following coding mode, macro block/image block that (i) will encode in this coding mode then selects to be used for INTRA coding mode that reference picture is encoded for the first regulation by " jump " signal, also namely in the situation that the reference picture of the macro block/image block that is encoded and (ii) will encodes without motion vector is encoded by means of the INTRA coding mode;
If the-the three regulation is used the INTER coding mode with affiliated motion vector for image block/macro block, then stipulate to select in the same manner the first regulation with the 3rd, wherein select in the same manner to be used for the motion vector of the 3rd regulation and the first INTER coding mode of stipulating.
13. according to the described method of one of aforementioned claim, wherein in image encoding apparatus (ENC) and in image decoding apparatus (DEC) synchronously the method between image encoding apparatus (ENC) and image decoding apparatus (DEC), notify following parameter one of at least:
-for the index (INX) of the image-region of the image area (BR) of reconstructed image (RBD) or reconstructed image (RBD), wherein said index (INX) comprises for the sign that identifies clearly the first regulation (V1) or the second regulation (V2);
-quantizing factor, it is used at least one image area is compressed in INTRA predictive mode or INTRA coding mode situation.
14. be used at least one image area (BR) of storage reconstructed image (RBD), described reconstructed image generates by image encoding apparatus (ENC) or by image decoding apparatus (DEC), with be used to image encoding apparatus (ENC) or image decoding apparatus (DEC) to provide at least one image area (BR) as with reference to the equipment (SV) of image (REF), have with lower device:
A) be used for stipulating that based on first (V1) is compressed at least one image area (BR) image compression unit (PC) of compressed data (KDAT);
B) for the memory cell (SRB) of depositing compressed data (KDAT);
C) be used for stipulating that based on second (V2) de-compresses into compressed data (KDAT) the image decompressor unit (PD) of at least one image area (BR);
It is characterized in that,
D) form the first regulation (V1) based on the 3rd regulation (V3), wherein the 3rd regulation (V3) defines the coding mode (M1) of image encoding apparatus (ENC),
E) form in the following manner the second regulation (V2), namely the second regulation (V2) is the 3rd contrary regulation (V3I) with the 3rd regulation (V3) contrary.
15. equipment according to claim 14 (SV), this equipment have processing unit (VMIT) in addition, described processing unit is constituted as at least one that makes it possible to achieve and implement according to claim 2 to one of the 13 described steps.
16. according to claim 1 to one of 13 described methods or according to claim 14 with one of 15 application of described equipment in Image Coding unit (ENC) and/or image decoding unit (DEC).
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| EP11150714 | 2011-01-12 | ||
| EP11150714.1 | 2011-01-12 | ||
| PCT/EP2012/050430 WO2012095487A1 (en) | 2011-01-12 | 2012-01-12 | Compression and decompression of reference images in a video coding device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103329527A true CN103329527A (en) | 2013-09-25 |
| CN103329527B CN103329527B (en) | 2017-03-01 |
Family
ID=45495941
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201280005271.5A Expired - Fee Related CN103329527B (en) | 2011-01-12 | 2012-01-12 | The compression of reference picture and decompression in video encoder |
| CN201280005268.3A Expired - Fee Related CN103283231B (en) | 2011-01-12 | 2012-01-12 | The compression and decompression of reference picture in video encoder |
Family Applications After (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201280005268.3A Expired - Fee Related CN103283231B (en) | 2011-01-12 | 2012-01-12 | The compression and decompression of reference picture in video encoder |
Country Status (6)
| Country | Link |
|---|---|
| US (2) | US9723318B2 (en) |
| EP (2) | EP2617198A1 (en) |
| JP (2) | JP5698382B2 (en) |
| KR (2) | KR101873207B1 (en) |
| CN (2) | CN103329527B (en) |
| WO (2) | WO2012095487A1 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108810556A (en) * | 2017-04-28 | 2018-11-13 | 炬芯（珠海）科技有限公司 | The method, apparatus and chip of compressed reference frame |
| CN109511008A (en) * | 2018-11-27 | 2019-03-22 | 成都索贝数码科技股份有限公司 | A method of it is stored based on object and supports that video and audio file content is additional |
Families Citing this family (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5698382B2 (en) | 2011-01-12 | 2015-04-08 | シーメンス アクチエンゲゼルシヤフトＳｉｅｍｅｎｓ Ａｋｔｉｅｎｇｅｓｅｌｌｓｃｈａｆｔ | Reference image compression and decompression method in video coder |
| EP3902258B1 (en) * | 2011-06-30 | 2025-03-26 | Telefonaktiebolaget LM Ericsson (publ) | Reference picture signaling |
| JP2014123841A (en) * | 2012-12-20 | 2014-07-03 | Jvc Kenwood Corp | Image encoder and image encoding method |
| US10015515B2 (en) * | 2013-06-21 | 2018-07-03 | Qualcomm Incorporated | Intra prediction from a predictive block |
| US9883197B2 (en) | 2014-01-09 | 2018-01-30 | Qualcomm Incorporated | Intra prediction of chroma blocks using the same vector |
| JP2015146474A (en) * | 2014-01-31 | 2015-08-13 | 株式会社アクセル | Encoding device and decoding device |
| JP5879571B2 (en) * | 2014-01-31 | 2016-03-08 | 株式会社アクセル | Encoding device and decoding device |
| US10419512B2 (en) * | 2015-07-27 | 2019-09-17 | Samsung Display Co., Ltd. | System and method of transmitting display data |
| KR20170017573A (en) * | 2015-08-07 | 2017-02-15 | 삼성전자주식회사 | Image Data Processing method and electronic device supporting the same |
| US10798396B2 (en) | 2015-12-08 | 2020-10-06 | Samsung Display Co., Ltd. | System and method for temporal differencing with variable complexity |
| EP3185556A1 (en) | 2015-12-21 | 2017-06-28 | Thomson Licensing | Method and apparatus for combined adaptive resolution and internal bit-depth increase coding |
| JP6875821B2 (en) * | 2016-10-19 | 2021-05-26 | キヤノン株式会社 | Image coding device, imaging device, image coding method and program |
| US10838922B2 (en) * | 2017-03-31 | 2020-11-17 | International Business Machines Corporation | Data compression by using cognitive created dictionaries |
| JP7359653B2 (en) * | 2019-11-06 | 2023-10-11 | ルネサスエレクトロニクス株式会社 | Video encoding device |
| KR102167561B1 (en) | 2020-03-11 | 2020-10-20 | 톈진 나가르 메커니컬 인더스트리 리미티드 컴퍼니 | High-pressure Plunger Type Single Diaphragm Pump |
| KR102167568B1 (en) | 2020-03-11 | 2020-10-20 | 톈진 나가르 메커니컬 인더스트리 리미티드 컴퍼니 | High-pressure Plunger Type Double Diaphragm Pump |
| US12088643B2 (en) * | 2022-04-15 | 2024-09-10 | Google Llc | Videoconferencing with reduced quality interruptions upon participant join |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5933195A (en) * | 1997-09-26 | 1999-08-03 | Sarnoff Corporation | Method and apparatus memory requirements for storing reference frames in a video decoder |
| US6229852B1 (en) * | 1998-10-26 | 2001-05-08 | Sony Corporation | Reduced-memory video decoder for compressed high-definition video data |
| US6470051B1 (en) * | 1999-01-25 | 2002-10-22 | International Business Machines Corporation | MPEG video decoder with integrated scaling and display functions |
| US20100098169A1 (en) * | 2008-10-16 | 2010-04-22 | Texas Instruments Incorporated | Method and apparatus for motion estimation using compressed reference frame |
| CN102215399A (en) * | 2010-04-09 | 2011-10-12 | 新港传播媒介公司 | Frame buffer compression method, video encoder and system of video encoder |
Family Cites Families (32)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3165296B2 (en) | 1992-12-25 | 2001-05-14 | 三菱電機株式会社 | Inter-frame coding processing method, inter-frame coding processing method, and coding control method |
| US6208689B1 (en) | 1996-03-04 | 2001-03-27 | Mitsubishi Denki Kabushiki Kaisha | Method and apparatus for digital image decoding |
| CA2185753C (en) * | 1996-03-04 | 2000-09-12 | Hideo Ohira | Digital image decoding apparatus |
| JP3575508B2 (en) | 1996-03-04 | 2004-10-13 | Ｋｄｄｉ株式会社 | Encoded video playback device |
| US5818530A (en) * | 1996-06-19 | 1998-10-06 | Thomson Consumer Electronics, Inc. | MPEG compatible decoder including a dual stage data reduction network |
| US5987180A (en) | 1997-09-26 | 1999-11-16 | Sarnoff Corporation | Multiple component compression encoder motion search method and apparatus |
| JP2000165875A (en) * | 1998-11-26 | 2000-06-16 | Oki Electric Ind Co Ltd | Moving image resolution converting and encoding/ decoding device with a little memory capacity |
| WO2000042772A1 (en) * | 1999-01-15 | 2000-07-20 | Koninklijke Philips Electronics N.V. | Coding and noise filtering an image sequence |
| US6983101B1 (en) | 1999-03-31 | 2006-01-03 | Sanyo Electric Co., Ltd. | Image recording apparatus, image reproducing apparatus and image recording/reproducing apparatus |
| JP3286615B2 (en) * | 1999-03-31 | 2002-05-27 | 三洋電機株式会社 | Video recording device, video reproducing device and video recording / reproducing device |
| JP2002077925A (en) * | 2000-08-31 | 2002-03-15 | Kddi Corp | Digital compressed video stream re-encoder |
| US6983017B2 (en) | 2001-08-20 | 2006-01-03 | Broadcom Corporation | Method and apparatus for implementing reduced memory mode for high-definition television |
| JP4209631B2 (en) * | 2002-05-23 | 2009-01-14 | パナソニック株式会社 | Encoding device, decoding device, and compression / decompression system |
| KR100654436B1 (en) * | 2004-07-07 | 2006-12-06 | 삼성전자주식회사 | Method for video encoding and decoding, and video encoder and decoder |
| CN101023673B (en) * | 2004-09-16 | 2011-06-08 | 汤姆逊许可证公司 | Video codec with weighted prediction utilizing local brightness variation |
| SG130962A1 (en) | 2005-09-16 | 2007-04-26 | St Microelectronics Asia | A method and system for adaptive pre-filtering for digital video signals |
| JP2007166323A (en) * | 2005-12-14 | 2007-06-28 | Matsushita Electric Ind Co Ltd | Image decoding apparatus and image decoding method |
| US7903733B2 (en) | 2006-01-26 | 2011-03-08 | Qualcomm Incorporated | Adaptive filtering to enhance video encoder performance |
| JP4384130B2 (en) * | 2006-03-28 | 2009-12-16 | 株式会社東芝 | Video decoding method and apparatus |
| WO2007114368A1 (en) | 2006-03-30 | 2007-10-11 | Kabushiki Kaisha Toshiba | Image coding apparatus and method, and image decoding apparatus and method |
| JP4973886B2 (en) * | 2006-06-21 | 2012-07-11 | 日本電気株式会社 | Moving picture decoding apparatus, decoded picture recording apparatus, method and program thereof |
| EP2041981B1 (en) * | 2006-07-18 | 2013-09-04 | Thomson Licensing | Methods and apparatus for adaptive reference filtering |
| US8503521B2 (en) * | 2007-01-16 | 2013-08-06 | Chih-Ta Star SUNG | Method of digital video reference frame compression |
| WO2009055898A1 (en) | 2007-11-02 | 2009-05-07 | Ecole De Technologie Superieure | Method and system for generating a quality prediction table for quality-aware transcoding of digital images |
| US8270739B2 (en) | 2007-12-03 | 2012-09-18 | Ecole De Technologie Superieure | System and method for quality-aware selection of parameters in transcoding of digital images |
| GB2457262A (en) | 2008-02-08 | 2009-08-12 | Linear Algebra Technologies | Compression / decompression of data blocks, applicable to video reference frames |
| US8548041B2 (en) * | 2008-09-25 | 2013-10-01 | Mediatek Inc. | Adaptive filter |
| JP2010098352A (en) * | 2008-10-14 | 2010-04-30 | Panasonic Corp | Image information encoder |
| CN102165778A (en) * | 2009-02-10 | 2011-08-24 | 松下电器产业株式会社 | Image processing device, image processing method, program and integrated circuit |
| US8724707B2 (en) * | 2009-05-07 | 2014-05-13 | Qualcomm Incorporated | Video decoding using temporally constrained spatial dependency |
| JP5361998B2 (en) * | 2009-05-29 | 2013-12-04 | 三菱電機株式会社 | Image encoding device, image decoding device, image encoding method, and image decoding method |
| JP5698382B2 (en) | 2011-01-12 | 2015-04-08 | シーメンス アクチエンゲゼルシヤフトＳｉｅｍｅｎｓ Ａｋｔｉｅｎｇｅｓｅｌｌｓｃｈａｆｔ | Reference image compression and decompression method in video coder |
-
2012
- 2012-01-12 JP JP2013548845A patent/JP5698382B2/en not_active Expired - Fee Related
- 2012-01-12 CN CN201280005271.5A patent/CN103329527B/en not_active Expired - Fee Related
- 2012-01-12 JP JP2013548842A patent/JP5710022B2/en not_active Expired - Fee Related
- 2012-01-12 KR KR1020137020898A patent/KR101873207B1/en not_active Expired - Fee Related
- 2012-01-12 EP EP12705230.6A patent/EP2617198A1/en not_active Withdrawn
- 2012-01-12 KR KR1020137020897A patent/KR101890203B1/en not_active Expired - Fee Related
- 2012-01-12 WO PCT/EP2012/050430 patent/WO2012095487A1/en active Application Filing
- 2012-01-12 CN CN201280005268.3A patent/CN103283231B/en not_active Expired - Fee Related
- 2012-01-12 WO PCT/EP2012/050435 patent/WO2012095490A1/en active Application Filing
- 2012-01-12 US US13/979,497 patent/US9723318B2/en active Active
- 2012-01-12 EP EP12700392.9A patent/EP2614647B1/en not_active Not-in-force
- 2012-01-12 US US13/979,601 patent/US9398292B2/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5933195A (en) * | 1997-09-26 | 1999-08-03 | Sarnoff Corporation | Method and apparatus memory requirements for storing reference frames in a video decoder |
| US6229852B1 (en) * | 1998-10-26 | 2001-05-08 | Sony Corporation | Reduced-memory video decoder for compressed high-definition video data |
| US6470051B1 (en) * | 1999-01-25 | 2002-10-22 | International Business Machines Corporation | MPEG video decoder with integrated scaling and display functions |
| US20100098169A1 (en) * | 2008-10-16 | 2010-04-22 | Texas Instruments Incorporated | Method and apparatus for motion estimation using compressed reference frame |
| CN102215399A (en) * | 2010-04-09 | 2011-10-12 | 新港传播媒介公司 | Frame buffer compression method, video encoder and system of video encoder |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108810556A (en) * | 2017-04-28 | 2018-11-13 | 炬芯（珠海）科技有限公司 | The method, apparatus and chip of compressed reference frame |
| CN109511008A (en) * | 2018-11-27 | 2019-03-22 | 成都索贝数码科技股份有限公司 | A method of it is stored based on object and supports that video and audio file content is additional |
| CN109511008B (en) * | 2018-11-27 | 2021-07-13 | 成都索贝数码科技股份有限公司 | Method for supporting video and audio file content addition based on object storage |
Also Published As
| Publication number | Publication date |
|---|---|
| US9398292B2 (en) | 2016-07-19 |
| US20130301711A1 (en) | 2013-11-14 |
| JP5710022B2 (en) | 2015-04-30 |
| US20140003517A1 (en) | 2014-01-02 |
| KR101890203B1 (en) | 2018-08-21 |
| US9723318B2 (en) | 2017-08-01 |
| JP2014506442A (en) | 2014-03-13 |
| KR101873207B1 (en) | 2018-07-02 |
| CN103283231B (en) | 2018-04-27 |
| JP5698382B2 (en) | 2015-04-08 |
| CN103329527B (en) | 2017-03-01 |
| EP2617198A1 (en) | 2013-07-24 |
| EP2614647B1 (en) | 2017-10-18 |
| KR20140042779A (en) | 2014-04-07 |
| CN103283231A (en) | 2013-09-04 |
| KR20140042778A (en) | 2014-04-07 |
| WO2012095490A1 (en) | 2012-07-19 |
| JP2014506441A (en) | 2014-03-13 |
| EP2614647A1 (en) | 2013-07-17 |
| WO2012095487A1 (en) | 2012-07-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103329527A (en) | Compression and decompression of reference images in a video coding device | |
| RU2694013C1 (en) | Improved intra prediction encoding using planar views | |
| JP6851265B2 (en) | A split block coding method for video coding, a split block decoding method for video decoding, and a recording medium that realizes this. | |
| TWI632805B (en) | Video decoding method, video encoding method, video decoding device, video encoding device, and non-transitory computer readable storage medium | |
| KR101538704B1 (en) | Method and apparatus for encoding and decoding an image using an interpolation filter adaptively | |
| CN102342101B (en) | Combined scheme for interpolation filtering, in-loop filtering and post-loop filtering in video coding | |
| KR101947657B1 (en) | Method and apparatus for encoding intra prediction information | |
| KR101543138B1 (en) | Video image encoding method and video image decoding method | |
| US8325797B2 (en) | System and method of reduced-temporal-resolution update for video coding and quality control | |
| WO2008004768A1 (en) | Image encoding/decoding method and apparatus | |
| KR20110071231A (en) | Encoding Method, Decoding Method and Apparatus | |
| KR20150027530A (en) | High efficiency video coding intra frame prediction apparatus and method thereof | |
| KR20070026914A (en) | A recording medium on which an image encoding and decoding apparatus, a method thereof, and a program for performing the same are recorded. | |
| KR20130045151A (en) | Method and apparatus for generating reconstruction block | |
| CN100401780C (en) | Method and system for generating a transform size syntax element for video decoding | |
| KR20130003707A (en) | Method and apparatus for image encoding/decoding using efficient non-fixed quantization | |
| Hollmann | Coding of Guiding Data for VideoTranscoding | |
| CN116980609A (en) | Video data processing method, device, storage medium and equipment | |
| KR20130050534A (en) | Methods of encoding using hadamard transform and apparatuses using the same | |
| KR20140038323A (en) | Apparatus and method for inter-layer reference of multi-layer video |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170301 Termination date: 20200112 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |